Airbnb is a website for people to list, find, and rent lodging. It deals between travelers and homeowners with spare rooms to rent in a short term and generated more than ten million overnight stays in 2012. Currently New York City is the largest Airbnb market in US and second largest in the world after Paris, majorly in Manhattan and Brooklyn.
This study aims to study the pricing dynamics of Airbnb units in the New York City area, and raise the questions such as: Are similar prices for units clustered together? Is there a geographical price trend in the New York City area? How are the prices for rooms or apartments decided? It may be obvious that certain locations are more popular for the Airbnb market, and it also might be obvious that crime rates negatively impact whether a customer will rent a room in that area. We attempted to quantify these characteristics. In our study we use a form of multivariate regression- the hedonic regression- to model different attributes affect’s on the price of an Airbnb unit and come to some conclusion about trends associated with our regression results.
Through data exploration and visualization it shows interesting geographical findings on Airbnb market in New York City. First, Manhattan and Brooklyn has the most of listings yet with very different listing characteristics regarding the listing price, location, room type and density. For instance, Brooklyn has more evenly distributed number of listings in entire home and private room, yet Manhattan has much more listing solely on entire home. The locations of listings in Manhattan are more spatially dispersed by different neighborhood, however most of listings in Brooklyn are highly concentrated in Williamsburg.
Figure above shows the top 10 listed neighborhood (in percentage) in Manhattan and Brooklyn.
Figure above shows where are the top 100 most expensive listings by count in Manhattan and Brooklyn.
Figure above shows difference of listing price range and distribution in Manhattan and Brooklyn.
Although with spatial agglomerations of listings, there is no strong spatial pattern regarding the listing price. The grouped data of listing based on postal code shows that the standard deviation of listing price within the same postal code is quite high. Especially, among top ten postal code zones with the highest price deviation, nine of them are in lower portion of Manhattan including Midtown, East Village, Hell’s Kitchen and SoHo. The fact of such big deviations in pricing reveals a wide range and variance in listing market spatially, which would be interesting for future study to understand the real-estate value and tourism economy.
Figure above shows top 10 neighborhoods with most deviated listing prices in Manhattan.